Projection: A Mechanism for Human-like Reasoning in Artificial Intelligence

TL;DR

The paper introduces 'projection', a top-down inference mechanism inspired by human reasoning, which enhances AI systems' ability to apply knowledge flexibly across diverse and challenging situations in vision, robotics, and language.

Contribution

It presents 'projection' as a novel inference mechanism that improves AI reasoning and knowledge application in varied contexts, addressing limitations of current AI systems.

Findings

Projection improves recognition in difficult visual conditions.

It enables AI to adapt knowledge to new, challenging situations.

Relevance to commonsense reasoning is discussed.

Abstract

Artificial Intelligence systems cannot yet match human abilities to apply knowledge to situations that vary from what they have been programmed for, or trained for. In visual object recognition methods of inference exploiting top-down information (from a model) have been shown to be effective for recognising entities in difficult conditions. Here this type of inference, called `projection', is shown to be a key mechanism to solve the problem of applying knowledge to varied or challenging situations, across a range of AI domains, such as vision, robotics, or language. Finally the relevance of projection to tackling the commonsense knowledge problem is discussed.